[01:11] New bug: #845247 in net-snmp (main) "SNMP mibs +all does not work in snmpd.conf" [Undecided,New] https://launchpad.net/bugs/845247 [02:28] gents is there a better way to hide a file/folder than placing the . infront of it. There is a shell discrepancy happening and i can't fix it as the incoming user (system) isn't within my control. Basically...if i log in locally to the box or get to it via the network (NFS or samba) i can't see the folder unless i use the -a. If the other user logs in via the network they can see it [02:28] with a straight ls. I would like this particular folder to be compeltely hidden from view for any incoming users. [02:33] lickalott: have them run "alias ls" [02:33] They probably have explicitly asked for it to be -a by default [02:34] And no, there is no better way, except possibly for something filesystem-specific (that I'm not aware of). [02:35] was afraid of that.... I've seen this before i.e. changing default shell from bash to csh or tcsh. Bash will require the -a and tcsh will show it regardless. [02:42] Sorry, I don't support csh/tcsh. [02:42] But I think it's probably still an alias, just that e.g. /etc/tcshrc sets it and /etc/bashrc doesn't [02:43] on the phone with one of my guys now having him check [02:43] thks! [02:44] lickalott: Log it from the network, and (a) check what shell you are running, and then (b) see what alias ls outputs. [02:46] You could just, you know, read the rc files === Usuario896 is now known as koolhead11 [03:06] just hung up the phone. said he "echo $SHELL" on the machine as the local user and it said csh but in etc/passwd is showed bash. He is accessing it via NFS with the user that i don't want to see the stuff. Had him vi .cshrc and strip out all aliases and try again. he's going to call me in a bit. [03:15] IMO best solution is to uninstall csh :-) [03:47] hi all === koolhead11 is now known as koolhead17 [03:57] I setup Ubuntu Cloud in test environment, and I'm having an SSL issue with the store. Anyone willing to help with this one:) === jussi01_ is now known as jussi [05:11] neoXsys: hey [05:11] koolhead17, Pong [05:11] how are you doing? [06:31] New bug: #845388 in postfix (main) "package postfix 2.8.2-1ubuntu2.1 failed to install/upgrade: subprocess installed post-installation script returned error exit status 75" [Undecided,New] https://launchpad.net/bugs/845388 [09:17] * Daviey curses Mr Bloom's Nursery. [09:50] Daviey: morning sir o/ [09:54] hey lynxman [09:56] Morning all [09:57] Daviey: should squid-deb-proxy-client work out of the box for installs of oneiric? [10:03] RoAkSoAx, GrueMaster: Notes and a level of automation of what I did yesterday re cobbler/arm/preseeds/sd cards etc... [10:03] http://pad.ubuntu.com/arm-server-netboot [10:06] jamespage: squid-deb-proxy-client worked so far for me, although your mileage will vary on the server if you don't use private networks, requires further tweaking [10:07] lynxman: do I need todo anything special? i.e. include flags in a preseed or suchlike? [10:07] jamespage: I did setup with mvo a couple debconf hooks on low prio for exactly that :) [10:07] jamespage: although tbh I can't recall the names right now :/ [10:09] lynxman: hmm - so they seem to be more about setup once installed [10:09] I want my installer to use it [10:09] jamespage: you can preset the debconf options before installing the package, that's what I did in orchestra v1.0 [10:09] hey! im making a file server for windows clients, what good to get the windows filesystem working with linux file systems? (Clients will be in different companies/locations so Samba isnt suitible cause it wants workgroups) [10:10] jamespage: should be easy to do through preseed shouldn't it [10:11] lynxman: one would hope so [10:11] jamespage: :) === sgo11_away is now known as sgo11 === jussi01_ is now known as jussi [10:15] Hi, with default installation of apache2 and ubuntu 11.04 natty. the ubuntu doc says ErrorDocument directive can be found in apache2.conf. But I failed to find it. The default 404 error output is also not from /usr/share/apache2/error/* from what I can see. I tried to grep -ir ErrorDocument under /etc/apache2/. no found. I know I can define this in virtual host config file. but I am just wondering which file defines ErrorDocument in default installation? [10:15] thanks alot. [10:33] http://httpd.apache.org/docs/2.0/mod/core.html#errordocument lists 4 options that Apache can be configured to handle Error Documents. If it is not defined, the default behaviour is to display a hardcoded error page. [10:40] just-a-visitor, ok. thanks a lot. I think it displays the hardcoded error page. no wonder I can not find it. [10:48] jamespage: It should.. yes.. but a number of people have reported woe [10:48] Daviey: not working so well for me either [10:49] :( [10:52] jamespage: Hmm, do you have squid-deb-proxy server component on your network? [10:52] Daviey: yes [10:53] advertises itself via avahi [10:53] I wonder if avahi doesn't work properly on arm [10:54] Daviey: might be - I'll try a x86 install instead [10:56] Daviey; hrm - might be a clue - base-installer: warning: /usr/lib/base-installer.d/80squid-deb-proxy-client returned error code 1 [10:58] Daviey: I had to enable is using early_command [10:58] and its still in universe so had to turn that on for udeb as well [10:59] oh [11:00] guess that might not be right? [11:00] it works fine if I install on the server after is been installed [11:01] i.e. squid-deb-proxy-client [11:10] jamespage: Need to check if it is seeded [12:06] New bug: #845536 in rabbitmq-server (main) "rabbitmq user showed as being logged in" [Undecided,New] https://launchpad.net/bugs/845536 [12:21] New bug: #845541 in solr (universe) "Remove libcommons-fileupload-java-universe from the Oneiric archive" [Undecided,New] https://launchpad.net/bugs/845541 [12:48] morning [13:17] How do you setup the mail server? [13:18] that's quite a wide question [13:18] Recently upgraded from 8.04.4 to 10.04.3. I'm experiencing a phenomena where a NIS slave will not bind to an NIS master if network-manager service is started. If I stop the service, suddenly NIS will bind and start. Anyone know the cause of this problem off-hand? [13:18] tp43: you need to work out what stlye of mail setup you want, then what software you need/want to do that [13:18] I am gonna install xmail and see what happens [13:18] Blinny: can you bind after you start network manager ? [13:18] ikonia: No. [13:18] tp43: Have you seen https://help.ubuntu.com/10.04/serverguide/C/email-services.html ? [13:19] Blinny: does network manage keep the same IP address as when you are not using network manager? [13:19] ikonia: Yes. [13:19] Blinny: so is network manager set for a static IP ? [13:19] ikonia: I can tell no difference in ifconfig or resolv.conf [13:19] ikonia: I have not historically used network-manager. However, the IPs of the NICs do not change. [13:20] Blinny: is network manager setup to use a static IP - or a dhcp ip [13:21] ikonia: I am from the Slackware 1.x world so I don't know much about network manager, just manual configuration using etc/network/interfaces . This is also a headless server so I don't do a lot of app stuff. Where are network manager's config files found? [13:21] Blinny: why are you using network manager ? [13:21] ikonia: Believe me, I don't want to! [13:21] Blinny: if you have put the details in network/interfaces it will conflict with network manager [13:21] Pici, thx [13:21] Blinny: ok, so disable it [13:21] ikonia: Brilliant. Can I purge? [13:22] Blinny: that's the issue, your probably creating a conflict [13:22] Blinny: if you're not wanting to use it, remove/disable it [13:22] I do have three other servers that have network-manager but don't have this NIS problem. I had thought I'd shelved/disabled. It's not listed in sysv-rc-conf but the change to upstart probably causing it to become active. [13:22] Is network-manager started by udev? What is the preferred method of disabling? [13:24] Blinny: is this machine running a desktop by any chance ? [13:24] I think upstart uses /etc/init/*.conf files. Ah after dbus. [13:24] ikonia: This is a LTSP machine, so yes. [13:24] ubuntu-desktop is installed. [13:24] oooh, I need an mta [13:25] tp43: I've had great luck with postfix. [13:25] tp43: YMMV [13:25] Blinny, hmm, I just went with exim4 [13:25] tp43: A good starting place. [13:25] Blinny: ok, so you may want to just disable it on the desktop, that's the easy way [13:26] ikonia: At the server console? [13:26] ikonia: Since NetworkManager seems to be The New Way To Go, I'm worried about purging it due to future dependencies. [13:26] Perhaps my worry is unfounded. [13:27] Blinny: why ? [13:28] there is no need to worry [13:28] So purge it is! I have a few laptops and I've found networkmanager nice for handling wireless but for normal servers, I'll be glad to see it go. [13:28] The package is network-manager? [13:30] network manager is great for a desktop, I think it's a really useful tool, especially for things such as multiple wireless networks, however if you're setting the info in /etc/network/interfaces, you're creating a conflict, use one or the other [13:31] ikonia: Thank you. Cheers mate. [13:32] Another NIS question: I have one master and three slave+clients. When the client couldn't bind to the master, it would not answer queries as a slave. I understand this is why a slave setup exists, so I'm assuming I have something wrong in my config. Does this sound right? [13:33] Wow, NIS? I haven't seen a NIS setup in a decade or so. [13:34] LDAP gave me nightmares about 6 years ago and I've never recovered. Some day I'll give another crack at 389 Dir. Server. [13:35] OK. network-manager removed on 3 servers. [13:35] Blinny: it depends if it is "current" in terms of it's maps [13:36] Does that require a normal start-up w/ bind, then connection lost? I.E. maps aren't cached across reboot? [13:39] jamespage: ping [13:39] hey zul [13:40] jamespage: so you know that bug you opened when running four lxc instances are you using nbd? [13:41] zul: maybe - is that something I would have set explicitly? I just installed nova-compute-lxc on an all-in-one install [13:42] no it should just modprobe nbd [13:42] jamespage: and which version are you using as well [13:42] of nova? [13:43] zul: I've torn down the system now but its was a fresh install on the date I reported the bug [13:44] * jamespage must remember to use ubuntu-bug [13:46] jamespage: okie dokie...im going to roll a new one today once it is in the archive can you retest because im running with -n 2 and i cant reproduce it [13:46] zul: OK - will [13:46] do [13:47] there was something a little odd about the way I setup the system when I saw that issue [13:47] I started with nova-compute-kvm [13:47] and then switched to nova-compute-lxc [13:47] not sure that would make a difference [13:47] jamespage: ah ok...yeah the configuration file is different [13:48] zul: well I gave it all a good kick before I tried to start any lxc instances [13:48] i.e. rebooted [14:32] Using Lucid, openswan fails to start on boot. One must manually do 'service openswan restart' (not start), I believe because of a locking issue, but may be related to using sysvinit rather than upstart. Anyone use openswan for ipsec or know of this issue? I may try to create /etc/init/openswan.conf and change the job to upstart to see if that helps, but this will be my first. [14:52] jamespage: I'm looking over your pad now. One thing I noticed is the mac address setting. Is that required for cobbler? Also, you shouldn't need to set the omapfb stuff for headless install. [14:54] GrueMaster: so cobbler uses it to set a specific netboot install target for the server - rather than picking up the generic netboot menu including x86 stuff [14:54] Ah. [14:54] I haven't used cobbler, so didn't know. [14:54] GrueMaster: if you can give me a better set of kernel options for the headless install much appreciated - I think I got those from here: [15:04] zul: I just setup my lxc nova all-in-one [15:05] when I terminate my lxc instances I get a load of qemu-nbd processes hanging around for them - which does not look right to me [15:05] jamespage: known bug...i have a branch that fixes it [15:05] zul: coolio [15:09] yay collectd-server ensemble fomula works [15:09] now for the web part [15:29] GrueMaster: (from a few days back) I might just be getting old, but I don't remember mentioning an IPv6 test suite at all - are you sure that was me? [15:30] GrueMaster: unless it was the tunnelbroker.net certification program [15:30] cjwatson: I never suggested that it was you. Others did. [15:31] The only thing I have seen for full ipv6 testing is from http://tahi.org, but it is software written for BSD. [15:32] Other than that, there are web sites that will check your ipv6 connectivity, but I feel that isn't enough to check the entire stack (dhcp6, dns6, etc). [15:33] ah, ok, well, I was replying to: [15:33] 19:19 cjwatson: Ping. Do you remember the name of the IPv6 test suite you mentioned at UDS? [15:34] I was told in other channels that you had said something. [15:34] GrueMaster: you should probably talk with stgraber if you aren't already, as he's working on some things along those lines [15:36] Might be easier just to send an email to ubuntu-dev or warthogs. I don't have time to bounce from one dev to another like this. [15:37] ( I think you were the third person I was pointed to). [15:40] is rssh still good to use to allow FTP and block SSH? [15:49] allow sftp? [15:50] ^ [16:12] Daviey: the dialog on -devel between CJ and CR may interest you folks [16:13] Daviey: basically, a chance of finding oneself with a server that does not pxe-boot on default is probably resolved [16:13] I'd like to understand why my init.d program (thin) start successfully from the console but not at boot time. Any clues to help me on that? [16:16] New bug: #845764 in libvirt (main) "continuous respawning of libvirt-bin with LXC/ARM/nova-compute" [Undecided,New] https://launchpad.net/bugs/845764 [16:18] \o/ [16:18] rDx3: $PATH may not be the same. In general, the environment will be different, but $PATH is probably the most likely cause. [16:19] zul: ^^ this rings a bell with something you saw during the openstack sprint [16:20] thank jmarsden, do you mean the path in the init.d file? I think it was created directly from thin server. how do I check that? [16:20] kim0 are you online?? [16:21] lajjr: yes I am [16:21] great! [16:21] I have a ?? [16:21] rDx3: I can't walk you through troubleshooting a script right now, no time. you can echo $PATH to see what the value of PATH is. [16:21] lajjr: shoot :) [16:21] rDx3: I'm about to leave for work... [16:22] jmarsden: ok thanks [16:22] I completed the security ones and there are done.. [16:22] lajjr: you are awesome :) [16:24] On the next ones can I combine those chapters as one for a view, they will still be separate. in real form in guide just in the review branch?? [16:25] lajjr: It's probably best to keep a review per chapter (not sub-chapter) .. Having one review include two full chapters might not be a good idea, since the reviewer might not be confident on both topics [16:26] OK great just checking I made a chapter 13 branch and a 14 branch just have to push in a little bit.. [16:28] lajjr: awesome :) keep me coming .. Thanks man, it's really appreciated [16:28] kim0, I got them and they are separate, just didn't want to make a mistake by putting them together. [16:28] * kim0 nods [16:29] np Thank You kim0 [16:32] jamespage: werid [16:33] zul: potential regression? [16:34] jamespage: could be i would have to double check [16:35] how do I know under witch user a startup script is run? [16:38] morning SpamapS [16:39] SpamapS: zookeeper in Debian now contains the Ubuntu changes you made - do you want to re-sync as general housekeeping or leave it until next release? [16:41] jamespage: let it sync next release. :) [16:41] jamespage: but thanks for the heads up :) [16:41] SpamapS: that was my thinking to :-) [16:41] np [16:44] hggdh: nice [16:51] jamespage: effing [16:51] zul: same thing or something new? [16:51] same thing [16:51] bah [16:51] im going to go nuts [18:02] I just installed ubuntu server 11.04 on a new machine and grub failed to install on the raid array. I have a 4 disk RAID1 for / and a 4 disk RAID6 for /home [18:02] the machine is unbootable, how do I use the rescue disk to boot it and try and install grub again? [18:22] New bug: #845871 in qemu-kvm (main) "qemu-common cannot be installed because of unsatisfied dependency" [Undecided,New] https://launchpad.net/bugs/845871 [18:26] eh? [18:34] hallyn: ping [18:37] RoAkSoAx: yo === zoopster is now known as jpugh === jpugh is now known as zoopster [19:53] Daviey, zul: How are those fixes for LXC coming along? [19:53] Openstack+LXC, I mean. [20:04] soren: i have a branch that fixes the shutdown issues, so its going pretty good [20:06] Perhaps someone may know here.. token-bucket filter documentation for minburst says "To calculate the maximum possible peakrate, multiply the configured mtu by 100 (or more correctly, HZ, which is 100 on intel, 1024 on Alpha)"... are they talking about bus speed or what? [20:09] soren: i have to buy a new chair tonight though so ill have more fixes tonight hopefully [20:11] zul: Are your patches on LP somewhere? [20:12] soren: https://code.launchpad.net/~zulcss/nova/nova-fix-lxc-shutdown [20:13] zul: ta [20:13] soren: no worries [20:16] Ah, nvm, is kernel ticks [21:00] glebihan: just to let you know that I finally solved the problem thanks for your precious directions [21:09] rDx3, great :) glad I could be of some help [21:29] hi guys [21:29] i can't start mysqld on ubuntu [21:29] no pid file, no sock file, but it either claims it's already started or gives me weird errors [21:29] any ideas? [21:32] mister_bubbles: What command are you using to try and start it? [21:33] eg: upstart method or sysvinit [21:35] service mysql start [21:35] and start mysql [21:36] and /etc/init.d/mysql start [21:37] mister_bubbles: If you use a particular method to start it, use that method to stop or restart it. Otherwise it gets confused because the one method has no record of the pid or sock file the other method uses [21:39] i've tried all the methods 5 times [21:39] none of them do anything :( [21:40] eg: if you use start then don't use /etc/init.d/ [21:44] hallyn, around ? [21:44] never mind [21:44] start .... stop means upstart knows is running, what it's pid and how to handle. /etc/init.d/ start ... /etc/init.d/ stop means sysvinit knows those things. But if you start/stop then: /etc/init.d/ start/stop then it won't work [21:47] i originally killed the job using 'kill ' [21:49] What would be a good way to edit the filesystem remotely from a win7 pc? root ftp isn't I hear :) [21:50] sms_: Putty ssh client for windows [21:51] but I can't copy files from the w7 with putty or can I? [21:52] I mean move files to the server [21:52] hi, how can I install ATI drivers on server edition please? This is for a laptop running server edition that will also run XBMC [21:53] sms_: Ah. You said "edit" :) For scp under windows you can use PSCP , same place you get PuTTY from [21:54] oo thanks :) [21:54] sms_: You're welcome [21:56] New bug: #846057 in cloud-init (main) "cloud-init on oneiric does not honor "ubuntu-pass=*"" [Undecided,New] https://launchpad.net/bugs/846057 [22:06] argh [22:06] my /etc/init.d/mysql start just hangs forever [22:06] any ideas? [22:12] anyone? [22:12] mister_bubbles: try starting mysqld manually, or read the log file [22:12] there's nothing in the log files [22:12] neither error.log nor syslog [22:12] just termination notice and respawn notice [22:13] and i tried starting it manually in every way known to me (/etc/init.d/mysql start, service mysql start, start mysql) [22:13] it just hangs there [22:13] or do you have another way of starting it manually? [22:14] hu, interesting. when I do just 'mysqld', I get: [22:14] mysqld [22:14] 110909 15:14:02 [Warning] The syntax '--log_slow_queries' is deprecated and will be removed in MySQL 7.0. Please use '--slow_query_log'/'--slow_query_log_file' instead. [22:14] 110909 15:14:02 [Warning] Can't create test file /mysql/mysql/ip-10-114-95-171.lower-test [22:14] 110909 15:14:02 [Warning] Can't create test file /mysql/mysql/ip-10-114-95-171.lower-test [22:14] mysqld: Can't change dir to '/mysql/mysql/' (Errcode: 2) [22:14] 110909 15:14:02 [ERROR] Aborting [22:14] 110909 15:14:02 [Note] mysqld: Shutdown complete [22:16] mister_bubbles: look at the init script and how it launches mysqld_safe [22:25] it does 'start mysql' [22:25] which gives me: start: Rejected send message, 1 matched rules; type="method_call", sender=":1.10" (uid=1000 pid=9616 comm="start) interface="com.ubuntu.Upstart0_6.Job" member="Start" error name="(unset)" requested_reply=0 destination="com.ubuntu.Upstart" (uid=0 pid=1 comm="/sbin/init")) [22:27] mister_bubbles: that's using the upstart script [22:27] mister_bubbles: I suggested starting it manually, but using the same method of launching mysqld_safe [22:29] well, how do i manually start it? [22:30] read the init script and see how it launches mysqld follow that through manually [22:30] that's what i just did [22:30] it uses start [22:31] when I type mysqld_safe, i get: 110909 15:30:25 mysqld_safe Logging to syslog. [22:31] 110909 15:30:25 mysqld_safe Starting mysqld daemon with databases from /mysql/mysql [22:31] 110909 15:30:26 mysqld_safe mysqld from pid file /mysql/mysql/ip-10-114-95-171.pid ended [22:31] but then its not running with ps aux [22:31] mister_bubbles: is your database in /mysql/mysql ? [22:31] i guess so, i don't know [22:31] mister_bubbles: checking is a good start..... [22:32] my.cnf -> datadir says so [22:32] is that the right one? [22:32] mister_bubbles: the data dir is /mysql ? [22:32] it is /mysql/mysql [22:32] mister_bubbles: so are your databases there ? [22:33] (ubuntu doesn't put them in /mysql/mysql by default so someones changed that) [22:39] there are some files there [22:39] where are they usually? [22:39] there's some ib_logfiles and ibdata and stuff [22:40] so there is some data there, who setup this machine ? [22:40] i don't know, somebody in the company i guess [22:41] mister_bubbles: look for some MYD or MYI files in there [22:43] no, nothing [22:43] theres debian-5.1.flag, ibdata1, ib_logfile0, ib_logfile1, mysql folder (empty) and mysql_upgrade_info [22:43] mister_bubbles: that's why it's not starting up [22:43] mister_bubbles: or at least one of the problems, no database files in there [22:44] hu [22:44] how are they called? myd? [22:44] maybe i shall find for themn [22:44] I've just told you the extensions [22:46] find / finds no such thing [22:46] neither myd nor myi [22:46] look at the case [22:46] ah [22:46] there are some in /mnt/mysql/msql [22:48] what an odd place [22:48] hm [22:48] well [22:49] i changed the path in my.cnf to /mnt/mysql/mysql, still doesn't start [22:49] mister_bubbles: what's the error this time [22:50] 110909 15:47:49 mysqld_safe Logging to syslog. [22:50] 110909 15:47:49 mysqld_safe Starting mysqld daemon with databases from /mnt/mysql/mysql [22:50] 110909 15:47:49 mysqld_safe mysqld from pid file /mnt/mysql/mysql/ip-10-114-95-171.pid ended [22:50] same as last time but with /mnt in front [22:50] hang on - please show me the output of ls -la /mnt/mysql/mysql [22:50] (use a pastebin) [22:52] http://pastebin.com/hMak0SUR [22:53] mister_bubbles: it's the wrong dir, thats the data dir for the mysql database specfically [22:53] mister_bubbles: try /mnt/mysql [22:53] mister_bubbles: the person who set this up needs shooting [22:54] http://pastebin.com/gCcEyhte [22:54] whys, whats wrong? [22:54] mister_bubbles: that's a better directory [22:57] so you have any ideas why it won't start? [22:58] paste the current error please. [22:58] well, when i do what? mysqld_safe? [22:59] 110909 15:59:05 mysqld_safe Logging to syslog. [22:59] 110909 15:59:05 mysqld_safe Starting mysqld daemon with databases from /mnt/mysql/mysql [22:59] 110909 15:59:05 mysqld_safe mysqld from pid file /mnt/mysql/mysql/ip-10-114-95-171.pid ended [23:00] mister_bubbles: I told you to change the directory [23:00] mister_bubbles: it should be /mnt/mysql [23:01] ah you mean trying it in the config [23:01] ok doing that now, sec [23:01] yay, it's running :D [23:01] thx [23:02] but the DB is kinda weird [23:02] i think it's the wrong one [23:02] that's up to the guy who maintained the server [23:02] but I'd shoot him for that mess [23:03] soren: There seems to have been some fixes landed in Essex already that haven't been proposed to Diablo. === [1]MarcelT3 is now known as MarcelT3 [23:36] New bug: #846104 in mysql-5.1 (main) "package mysql-client-5.1 (not installed) failed to install/upgrade: trying to overwrite '/usr/bin/mysql_client_test', which is also in package mysql-bench 5.0.92-2" [Undecided,New] https://launchpad.net/bugs/846104